Cardiovascular disease (CVD) is the number one killer of American men and women, and its economic burden is substantial and on the rise. Adults with depression are at elevated risk of CVD events and poor CVD prognosis. Unfortunately, past trials of depression treatments have not observed the anticipated cardiovascular benefits. A novel explanation for these null results is that the interventions in these trials, which all involved patients with preexisting CVD, were delivered too late in the natural history of CVD. To begin to evaluate our hypothesis that treating depression before clinical CVD onset could reduce CVD risk, we conducted an 8-year follow-up study of a positive depression trial, the IMPACT trial. Intervention patients without baseline CVD had a 48% lower risk of fatal/nonfatal myocardial infarction or stroke than did usual care patients. Although these results support our hypothesis, they are post hoc and observational, CVD outcomes were not pre-specified endpoints, and minimal mechanistic data was collected. To address the current need for a well-powered, prospective trial, we propose a Phase 2 randomized controlled trial of 216 primary care patients (=50 years, 60% female, 40% African American, 5% Latino) with a depressive disorder and elevated CVD risk but no clinical CVD. Patients will be randomized to one year of eIMPACT, our modernized version of the IMPACT intervention, or usual care. eIMPACT is a collaborative stepped care intervention involving a multidisciplinary team delivering evidenced-based depression treatments consistent with patient preference. We will modernize our intervention by incorporating computerized cognitive-behavioral therapy and delivering other treatment components via telephone. Our central hypothesis is that eIMPACT will improve endothelial dysfunction, which is considered a barometer of CVD risk, in depressed adults by decreasing autonomic dysfunction, systemic inflammation, and platelet activation. We will test our central hypothesis by carrying out these specific aims: (1) to determine whether eIMPACT reduces the excess CVD risk of depressed patients (primary outcome: endothelial dysfunction; exploratory outcome: incident CVD events) and (2) to examine candidate mechanisms underlying the effect of eIMPACT on CVD risk (secondary outcomes: markers of autonomic dysfunction, systemic inflammation, and platelet activation). A positive trial would generate the mechanistic rationale, efficacy evidence, and effect size estimates that are needed to justify and design a multisite, event-driven Phase 3 trial to confirm eIMPACT's efficacy in reducing CVD risk. Demonstrating that depression treatment reduces CVD risk, the primary expected outcome of this line of research, would have a substantial positive impact. It would identify a novel treatment target (depression) for CVD prevention efforts, and it would equip providers with a new disseminable and scalable tool (eIMPACT) to simultaneously treat depression and manage CVD risk of a large cohort of high-risk patients. Collectively, these changes to clinical practice should translate into reduced CVD morbidity, mortality, and costs.